Implication of the miR-184 and miR-204 Competitive RNA Network in Control of Mouse Secondary Cataract
- 91 Downloads
The high recurrence rate of secondary cataract (SC) is caused by the intrinsic differentiation activity of residual lens epithelial cells after extra-capsular lens removal. The objective of this study was to identify changes in the microRNA (miRNA) expression profile during mouse SC formation and to selectively manipulate miRNA expression for potential therapeutic intervention. To model SC, mouse cataract surgery was performed and temporal changes in the miRNA expression pattern were determined by microarray analysis. To study the potential SC counterregulative effect of miRNAs, a lens capsular bag in vitro model was used. Within the first 3 wks after cataract surgery, microarray analysis demonstrated SC-associated expression pattern changes of 55 miRNAs. Of the identified miRNAs, miR-184 and miR-204 were chosen for further investigations. Manipulation of miRNA expression by the miR-184 inhibitor (anti-miR-184) and the precursor miRNA for miR-204 (pre-miR-204) attenuated SC-associated expansion and migration of lens epithelial cells and signs of epithelial to mesenchymal transition such as α-smooth muscle actin expression. In addition, pre-miR-204 attenuated SC-associated expression of the transcription factor Meis homeobox 2 (MEIS2). Examination of miRNA target binding sites for miR-184 and miR-204 revealed an extensive range of predicted target mRNA sequences that were also a target to a complex network of other SC-associated miRNAs with possible opposing functions. The identification of the SC-specific miRNA expression pattern together with the observed in vitro attenuation of SC by anti-miR-184 and pre-miR-204 suggest that miR-184 and miR-204 play a significant role in the control of SC formation in mice that is most likely regulated by a complex competitive RNA network.
Cataracts are a major ophthalmologic concern, with an incidence throughout the population (1,2). Cataracts can result from eye injury through trauma, exposure to sunlight and a variety of age-related physiological manifestations including inflammatory diseases, diabetes and genetic predisposition (3, 4, 5, 6). Current cataract therapies include surgical extracapsular lens fiber removal and synthetic lens implantation that can lead to secondary cataract (SC), also known as posterior capsular opacification in humans. In general, SC etiology includes the transdifferentiation of anterior capsule residual lens epithelial cells into mesenchymal myofibroblast cells (epithelial-mesenchymal transition [EMT]) that can migrate and expand into the posterior area of the lens capsule. The corresponding lens opacity results from EMT-associated changes in crystallin proteins, upregulation of cytoskeletal proteins such as a smooth muscle actin (α-SMA) and fibrotic extracellular matrix remodeling (7, 8, 9). In the pursuit of novel prevention and postsurgical therapies, numerous studies have focused on analyzing the etiology of SC formation, looking at the genetic predisposition and epigenetics as well as genomic and proteomic gene expression patterns (5,10,11). To study the detailed mechanism of SC, rodent cataract surgery models were successfully used, with lens epithelial cells undergoing SC during the initial days after lens fiber removal (12, 13, 14). Recently, we suggested that microRNA (miRNA)-dependent post-transcriptional regulation of lens development-associated genes might also play a role in lens regeneration (15). MicroRNAs are small proportional 22-nucleotide-long noncoding RNAs that regulate mRNA breakdown or translational interference of tissue-specific genes expressed during development, proliferation, differentiation and cell death mechanisms (16,17). RNA interference therapy was proposed as a therapeutic tool for a variety of clinical conditions (18, 19, 20). A recent study by Park et al. (21) proposed that specific targeting of SC-associated regulatory factors such as nuclear factor-κB (NF-κB) by RNA interference might provide a promising tool for postsurgical prevention of SC-associated EMT. Correspondingly, this study uses a previously established mouse cataract surgery model for SC (12), microarray hybridization technology (22) and lens capsular bag culture to identify the miRNA expression pattern after cataract surgery and to determine the regulatory role of selective miRNAs, in particular miR-184 and miR-204, on SC etiology.
Materials and Methods
Mouse Cataract Surgery
Cataract surgery was performed according to Suetsugu-Maki et al. and Medvedovic et al. (12,13,22) with modifications. C57BL/6 mice (8 wks old, female; The Jackson Laboratory, Bar Harbor, ME, USA) were anesthetized by intraperitoneal injections of ketamine and xylazine (95 and 14.3 mg/kg, respectively). For analgesia, buprenorphine (1 mg/kg) was give subcutaneously. The cornea was incised, and anterior capsulectomy was performed by removal of the lens core and fiber cells from the lens capsular bag. Capsular bags were washed with 1× phosphate-buffered saline (PBS) containing Mg2+ and Ca2+ to remove residual fiber cells. After 1, 2 and 3 wks of cataract surgery, mice were euthanized and eyeballs were removed to study SC formation. The posterior eyeballs were cut off and lens capsules were separated from the inside of the eyeball. For the control, intact lens capsules were collected from intact mouse eyes. The lens capsules were incubated in RNAlater® (Ambion, Austin, TX, USA) at 4°C overnight, followed by storage of lens capsules at −80°C.
Lens tissue was homogenized in TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) for total RNA isolation according to Trask et al. (23). Total RNA purity, quantity and quality were determined using a NanoDrop spectrophotometer ND-1000 (Thermo Scientific, Waltham, MA, USA) and an Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA). The miRNA was isolated from the total RNA preparation using the Flash Page™ gel system (Ambion). The purified miRNA was amplified using the NCode™ miRNA Amplification System (Invitrogen) from which the sense strand RNA was isolated using the PureLink™ Micro Kit (Invitrogen). The sense strand miRNA was labeled with cyanine-3 and used for hybridization on mouse miRNA microarray 8 × 15K slide version 2 containing probes for 627 mouse and 39 mouse γ herpes virus miRNAs from the Sanger database v 12.0 (Agilent Technologies). Agilent Feature Extraction software (24) depicted 55 miRNA probes that were detected on at least one out of four arrays by determination of the signal-to-error ratio. Intensity values for each individual time point were adjusted by setting negative values to “1.” Data are based on single values. Hierarchical clustering of mean centered intensities was employed using a Euclidean distance measure to generate a heat map for the 55 detectable probes using the open source software Cluster/Treeview. The software was written by Michael Eisen while at Stanford University.
Capsular Bag Culture
A radial incision was made at the border of the cornea and the sclera using a scalpel and scissors followed by lens extrusion using forceps. Lenses were washed in Dulbecco’s modified Eagle medium/F12 (DMEM/F12, 1× antibiotics/antimycotics; Corning® cellgro®, Mediatech Inc., Manassas, VA, USA), and any residual tissue on the outer lens capsule was removed supported by incubation in 0.25% trypsin/EDTA (Cellgro) for 1–5 min. After submersion in DMEM/F12 supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich, St. Louis, MO, USA), the posterior capsule was positioned on a ring support structure at the base of the dish and the anterior capsule was opened by making three clockwise incisions, with incisions not extending the lens equator by keeping the lens positioned using forceps. The capsule was peeled from the lens fiber cell mass using small scissors, and the posterior capsule was pinned with the exterior facing the bottom of a 3-mm culture dish and the anterior edge forming a cuplike structure (capsular bag) using six entomological pins (D1; Watkins and Doncaster, Kent, UK). Residual lens fibers were removed by changing the medium to DMEM/F12 without FBS.
miRNA Target Prediction
Target binding sites for miR-184 and miR-204 were analyzed according to the current TargetScan (Bin3, Runx2, Meis2 [listed below]) and PicTar (Meis2 [listed below]) prediction programs. Regarding Mus musculus (mmu) miRNAs, TargetScan predicted conserved targets for mmu-miR-184 (18 conserved targets) and mmu-miR-204 (322 conserved targets) were screened to identify potential SC-associated genes listed by the AmiGO gene ontology database (http://amigo.geneontology.org) under the GO terms GO:0001837: epithelial to mesenchymal transition (http://amigo.geneontology.org/cgi-bin/amigo/term_details?term=GO:0001837&session_id=4145amigo1314025373) and GO:0002088: lens development in camera-type eye (http://amigo.geneontology.org/cgi-bin/amigo/term_details?term=GO:0002088&session_id=4145amigo1314025373). The identified genes were further analyzed for microarray-identified SC-associated miRNA binding sites using TargetScan accessed August 2011.
TargetScan predicted targets for mmu-miR-184 can be found at: http://www.targetscan.org/cgi-bin/targetscan/vert_50/targetscan.cgi?mirg=mmu-miR-184. TargetScan predicted targets for mmu-miR-204 can be found at: http://www.targetscan.org/cgi-bin/targetscan/vert_50/targetscan.cgi?species=Mouse&gid=&mir_c=&mir_sc=miR-204/211&mir_nc=&mirg=&sortType=cs&incl_nc=0. TargetScan 3′ untranslated regions (UTR) for Bin3 can be found at: http://www.targetscan.org/cgi-bin/targetscan/vert_50/view_gene.cgi?taxid=10090&gs=BIN3&showcnc=0&shownc=0. TargetScan 3′UTR for Runx2 can be found at: http://www.targetscan.org/cgi-bin/targetscan/vert_50/view_gene.cgi?taxid=10090&gs=RUNX2&showcnc=0&shownc=0#miR-204/211. TargetScan 3′UTR for Meis2 can be found at: http://www.targetscan.org/cgi-bin/targetscan/vert_50/view_gene.cgi?taxid=10090&gs=MEIS2&showcnc=0&shownc=0. To access PicTar 3′UTR for Meis2, use: http://pictar.mdc-berlin.de/cgi-bin/PicTar_vertebrate.cgi (search terms: has-miR-204, gene ID NM_172315).
Transfection of Capsular Bags
Capsular bag cultures were transfected using lipofectamine. Optimal transfection efficiency for miRNA constructs was determined before experiments by using the pMIR-REPORT luciferase system (Ambion). In brief, the pMIR-REPORT luciferase plasmid expressing luciferase mRNA fused to the antisense sequence for hsa-mir-148a cloned in the Spe1, and the HindIII multiple cloning site was transfected alone (L-anti-miR) or cotransfected with hsa-miR-148a (L-anti-miR plus miR-148a). Firefly luciferase activity was reported from cell lysates 48 h after transfection according to the Promega Luciferase Assay System by detecting counts per minute (cpm) using a scintillation counter as previously published (15) (Figure 2). Cell lysates from nontransfected cells were used to measure background cpm (control). Samples were done in triplicate, and each sample included the average of 20 cpm measurements. Data were analyzed by unpaired Student t test, and P < 0.05 was used as a criterion for significance. Transfection efficiency of cy3-labeled anti-miR control (AM17011; Ambion) at different concentrations (for example, 5, 50 and 500 nmol/L) with or without lipofectamine was determined after 24 and 48 h of transfection by counting the number of cy3-labeled miRNAs localized within cell bodies (Figure 2). Human miR-184 inhibitor (anti-miR-184, AM10207), anti-miR control (AM17011), precursor miRNA for miR-204 (pre-miR-204, PM11116) and pre-miR control (AM17120), respectively, were mixed under optimized transfection conditions at a concentration of 50 µmol/L with 125 µL serum-free DMEM/F12 for 15 min and combined with the lipofectamine mixture under short vortexing and incubation for another 15–30 min. Proportional three to four capsular bag explants per 3-mm Petri dish were supplemented with serum-free DMEM/F12 and transfected with the corresponding mixture.
Migration and Cell Expansion
Imaging was performed using a TS100 microscope (Nikon, Tokyo, Japan) under a phase-contrast setting with a charged-coupled device (CCD) camera (CoolSNAP cf2; Photometrics, Tucson, AZ, USA) and imaging software (Metamorph; Molecular Devices, Eugene, OR, USA). Percent cell confluence after capsular miRNA transfection was quantified from two different donor eyes after 24 and 48 h. The data were derived by dividing each capsular bag into four quadrants followed by measuring the average area of cells that migrated to the denuded area. Data were analyzed by two-way analysis of variance (ANOVA) by entering the standard deviation of two donor eyes with four measurements each (representing four quadrants). P < 0.05 was used as a criterion for significance.
Imaging was performed using a TS100 microscope (Nikon, Tokyo, Japan) under a fluorescence setting with a CCD camera (CoolSnapcf2) and imaging software (Metamorph). After the different transfection time periods, capsular bags were washed twice with PBS, followed by fixation in 2% paraformaldehyde in PBS, 0.1% Triton X-100 for 30 min and staining with primary mouse monoclonal antibody α-SMA 1:250 (Sigma-Aldrich). Integrated optical density of α-SMA expression after capsular bag transfection with miRNA constructs was quantified from two donor eyes (anti-miRNA constructs) and three donor eyes (pre-miRNA constructs) after 24, 48 and 72 h, respectively. The data were derived by dividing each capsular bag into four quadrants followed by measuring the integrated optical density of same-size panes in two different capsular locations (for example, the capsular edge and migration border). Data of anti-miRNA constructs were analyzed by two-way ANOVA by entering the standard deviation of two donor eyes with four measurements each (representing four quadrants). Data for pre-miRNA constructs were analyzed by unpaired Student t test and two-way ANOVA by entering the standard error from three donor eyes with four measurements each (representing four quadrants). P < 0.05 was used as a criterion for significance.
After lens extrusion, capsular bag culture was established by opening the posterior part of the lens capsule and pinning the anterior capsule with cells facing the interior of the capsular bag to preserve the maximum amount of lens epithelial cells for miRNA transfection. A total of four capsular bags were used per protein sample. To preserve all the cells contained in a sample, the growth medium containing supernatant was transferred to an Eppendorf tube. Capsular bag attached epithelial cells were detached by rigorous flushing with 750 µL PBS, followed by transfer of cells to the supernatant-containing tube. Proteins were extracted according to the NE-PER Nuclear and Cytoplasmic Extraction Kit (Pierce Biotechnology, Rockford, IL, USA). In brief, after centrifugation at 2,000g and removal of the supernatant, cells were taken up in 25 µL Cytoplasmic Extraction Reagent I (CER I), vortexed and incubated on ice for 10 min. After addition of CER II, samples were vortexed, incubated on ice for 1 min, vortexed and centrifuged at 13,000g for 5 min. The cytoplasmic supernatant was separated from the nuclear pellet. The nuclear pellet was dissolved in 15 µL Nuclear Extraction Reagent (NER), vortexed for 10 min and centrifuged at 13,000g for 5 min. The supernatant containing the nuclear fraction was pooled with the cytosolic fraction to obtain total cellular protein extracts.
Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) and Western Blotting
The complete protein samples derived from four lenses each were separated on 10% sodium dodecyl sulfate-polyacrylamide gels and transferred to polyvinylidene fluoride (PVDF) membranes. PVDF membranes were incubated with the primary polyclonal antibodies for bridging integrator 3 (BIN3) (ProteinTech Group, Chicago, IL, USA), Meis homeobox 2 (MEIS2) (Abcam, Cambridge, MA, USA), runt-related transcription factor 2 (RUNX2), β-actin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (loading control) (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Staining with primary antibodies was followed by fluorescein-linked secondary antibody at 1:600 dilution and tertiary antibody anti-fluorescein AP conjugate 1:1,000 according to the enhanced chemifluorescence Western Blotting Kit (Amersham, Piscataway, NY, USA), and membranes were scanned with a Biospectrum 500 Imaging System with an LM26 and BioChemi 500 Camera f/1.2 and Vision Works LS software. Antibody staining intensities were measured using ImageGauge 4.1 (Fuji Photo Film). Three samples per group with four lenses per sample were analyzed. Intensities of primary antibody staining were normalized to the β-actin or GAPDH loading controls through division. For comparison of controls versus miRNA treatments, a two-tailed unpaired Student t test was used. P < 0.05 was used as a criterion for significance.
All animal studies were approved by the University of Dayton Laboratory Animal Institutional Review Board.
All supplementary materials are available online at www.molmed.org .
Cataract Surgery for miRNA Expression Profiling
Previously, we demonstrated that SC can be modeled in mice by surgical removal of lens fibers from the capsular bag (12). The resulting lens regeneration after extra-capsular lens fiber removal consists of two critical phases. The initial phase is during the first week of regeneration and is marked by upregulation of EMT-specific extracellular matrix components, cytoskeletal proteins such as α-SMA and downregulation of crystallins. The second phase begins at wks 2 and 3 and is marked by an upregulation of lens structural proteins and lens differentiation (12,13). For the present study, 8-wk-old C57BL/6 mice underwent cataract surgery by capsulectomy, as previously described (12,13). Following 1, 2 and 3 wks after surgery, RNA was isolated from regenerating lenses to determine the miRNA expression profiles during these different time points.
Cataract Surgery Induces SC-Dependent Expression of miRNAs
Capsular Bag Culture Using Anti-miR-184 and Pre-miR-204
The effect of miRNAs on SC etiology was studied in a capsular bag culture model that allows efficient manipulation of selective miRNA expression with faster outcomes compared with an animal surgery model and has been previously used by our group and others (13,44). SC characteristics can be observed within the first 3 d of capsular bag culture. When looking at miRNAs that demonstrate abundant expression in lens with a potential regulatory role in SC, miR-184 and miR-204 represent promising candidates (25). For instance, both of these miRNAs exhibit differential expression patterns during lens differentiation, regeneration and cataract etiology (25,27,45,46). In addition, these two miRNAs were previously identified to possess binding sites to the 3′UTRs of potential SC target mRNAs; for example, miR-184 targets the cataract-associated GTPase binding protein bridging integrator 3 [Bin3], and miR-204 targets the homeobox transcription factor Meis2 and canonical Wnt signaling associated transcription factor Runx2, respectively (29,47, 48, 49). In accordance with the SC observed opposing expression pattern of miR-204/211 versus miR-184 and our previous observations that demonstrated differential expression of miR-184 and miR-204 during iris cell dedifferentiation and lens regeneration in the adult newt (27), both of these miRNAs were selected for further investigation. To inhibit potential miR-184-dependent effects on SC formation that might result from the observed upregulation of miR-184 after cataract surgery, the miRNA inhibitor anti-miR-184 and the corresponding anti-miR control were chosen. Equivalent to the observed downregulation of miR-204/211 cluster after cataract surgery and with regard to the close sequence homology between miR-204 and miR-211, we simulated an overexpression of the miR204/211 cluster by transfection of pre-miR-204 precursor miRNA and the corresponding pre-miR control. Before miRNA transfection of capsular bag cultures, transfection efficiency was confirmed by microscopy and luciferase assay (see Figure 2 for detailed information).
Attenuation of SC-Associated Cell Migration and Expansion by Anti-miR-184
Attenuation of SC-Associated Cell Migration and Expansion by Pre-miR-204
Transfection with pre-miR control demonstrated a SC characteristic appearance of LE cell migration and expansion to the middle of capsular bags with 23.07% ± 0.34% cell confluence at d 1 and 94.66% ± 2.74% confluence at d 2 (see Figure 3), resulting in complete confluence at d 3 (data not shown). Transfection with pre-miR-204 was comparable to observations with anti-miR-184, including a significant attenuation of LE cell migration and expansion (P < 0.0001, two-way ANOVA) to 5.66% ± 1.32% confluence of capsular bags at d 1 and 49.74% ± 0.72% confluence at d 2.
Attenuation of SC-Associated α-SMA Expression by Anti-miR-184
Attenuation of SC-Associated α-SMA Expression by Pre-miR-204
Similarly, 3-d transfection of capsular bags with either pre-miR-204 or pre-miR control demonstrated an increasing α-SMA expression at the capsular edges and migration borders (compare Figure 4 and Figure 5B). In comparison to the control, pre-miR-204 demonstrated an attenuation of α-SMA expression over the 3-d time period at the capsular edge with an IOD of 3.52 ± 0.44 at d 1 (versus 5.8 ± 1.24, pre-miR control) and 10.2 ± 0.66 at d 3 (versus 9.3 ± 0.38, pre-miR control) (P < 0.05, two-way ANOVA) (Figure 5B). Because capsular bags transfected with pre-miR-204 had reached complete confluence at d 3, the two-way ANOVA did not yield any conclusive results regarding temporal changes of α-SMA IOD over the 3-d time period. Thus, a significant attenuation of ±-SMA expression could be observed at the migration border after 2 d transfection with pre-miR-204 with an IOD of 1.23 ± 0.07 versus 2.65 ± 0.5 (pre-miR control) (P < 0.05, Student t test).
Regulation of SC-Associated Target Expression by Anti-miR-184
One of the predicted miR-184 targets includes the mRNA for GTPase Bin3. The selective knockout of Bin3 was found to promote cataract formation in Bin3 knockout mice (29,50). Correspondingly, we hypothesized that the presence of anti-miR-184 might limit a potential miR-184-dependent downregulation of BIN3 expression during SC. Unexpectedly, Western blotting analysis demonstrated no changes in BIN3 protein expression after 1, 3 and 18 h capsular bag transfection with anti-miR-184 (data not shown).
Regulation of SC-Associated Target Expression by Pre-miR-204
Besides Meis2, miR-204 was predicted to regulate other EMT-associated target mRNAs such as Smad4 and cell cycle regulators Cdc7, Cdc25b, E2f3 and Runx2 (48,54,55). According to our previous study that demonstrated an attenuation of SC-associated Runx2 mRNA expression in the presence of complement receptor C5 antagonist after 1 wk of cataract surgery, Runx2 was chosen as an additional target for pre-miR-204 treatment (13). In contrast to an expected attenuation of RUNX2, no changes in RUNX2 protein levels could be detected in the presence of miR-204 at the different time points of 1, 3 and 18 h of capsular bag culture (data not shown).
Bioinformatic Analysis of MiR-184 and MiR-204 Target 3′UTR Reveals Multiple Binding Sites of Other Competitive miRNAs
In summary, the results conclude that miR-184 and miR-204 play a significant role in the control of SC etiology in mice and are most likely regulated through a complex networking with other miRNAs.
Role of miRNAs in Attenuation of SC
Posttranscriptional RNA interference through miRNA or small interfering RNA (siRNA) represents a new therapeutic tool for treatment of cancer, cardiovascular disease, fibrosis and antiviral therapy (56, 57, 58, 59). Recently, the group of Park et al. further confirmed the suitability of RNA interference therapy for treatment of SC (21). Accordingly, this study focused on identifying the SC-associated miRNA expression pattern after cataract surgery using microarray technology. The identified miRNA expression pattern included 55 miRNAs with different ontologies, for example, regulation of cell growth, differentiation and development, EMT-dependent fibrosis and lens development and cataract. Interestingly, when looking at changes in lens-specific miRNAs, miR-184 demonstrated a SC-associated upregulation in comparison to an observed downregulation of the miR-204/211 cluster (25). These miRNAs have also been identified to play a role during iris cell dedifferentiation and lens regeneration in the adult newt, supporting a potential lens-regenerative function (27). Having in mind the potential therapeutic role of miRNAs in prevention of SC, these two selected miRNAs have predicted binding sites to the 3′UTR of cataract-associated target mRNAs, for example, the GTPase binding protein Bin3, the homeobox transcription factor Meis2 and canonical Wnt signaling associated transcription factor Runx2, respectively (29,47,48). Correspondingly, miRNA-as-sociated counterregulation of these EMT-associated target mRNAs might prevent formation of SC.
Evidence of SC Attenuation by Anti-miR-184 and Pre-miR-204
We found clear evidence that both anti-miR-184 and pre-miR-204 can attenuate SC within the capsular bag model, supporting the SC regulatory role of these two miRNAs. Both of these miR-NAs were able to attenuate anterior LE cell migration and expansion from the capsular edge into the middle of the posterior capsular bag area with a proportional 3.5-/4-fold decrease of cell confluence at d 1 and 1.8-/1.9-fold decrease at d 2. This result is further supported by attenuation of ±-SMA expression with a proportional 1.4-fold downregulation of ±-SMA IOD at the capsular edge and a proportional twofold downregulation of ±-SMA IOD at the migration border over the 3-d time period in the presence of anti-miR-184 or pre-miR-204.
Bioinformatic Analysis of miR-184 Target 3′UTR Reveals Multiple Binding Sites of Other Competitive miRNAs
One explanation for the insufficient regulation of BIN3 expression by anti-miR-184 targeting is the existence of other competitive miRNA binding sites within the 3′UTR of Bin3 mRNA. For instance, TargetScan (http://targetscan.org) predictions identify a binding site for miR-184 at the position 264–270 of the Bin3 3′UTR, thus also predicting a binding site for the let-7/98 cluster at position 348–354 (compare the Bin3 Web address in the Materials and Methods section and Figure 7A). Because of the observed upregulation of most let-7/98 cluster members, the intended miR-184 attenuation by anti-miR-184 might not be sufficient in preventing a potential SC-associated downregulation of BIN3.
Including Bin3 TargetScan predicted a total of 18 mmu-miR-184-specific target mRNAs, suggesting that the observed SC attenuation might also include miR-184-dependent regulation and targeting of other mRNAs (compare to the mmu-miR-184 Web address in Materials and Methods).
Bioinformatic Analysis of miR-204 Target 3′UTR Reveals Multiple Binding Sites of Other Competitive miRNAs
Besides binding of miR-204/211 at position 3770–3776 within the 3′UTR of Runx2, TargetScan predictions identify multiple other miRNA binding sites, including the mir23a/b cluster at position 996–1002 and two miR-30c/d clusters at position 3445–3451 and position 3456–3462 (compare the Runx2 Web addresses in Materials and Methods and Figure 7B). In accordance with the SC-associated up-regulation of miR-30d and the miR-23a/b cluster, the pre-miR-204-dependent effect on Runx2 target mRNA expression might be rendered ineffective.
Similarly, the PicTar Web interface (http://pictar.mdc-berlin.de) predicts multiple other miRNA binding sites for the Meis2 3′UTR besides miR-204/211, including SC upregulated miR-23a/b, miR-27a and the let-7/98 cluster, as well as miR-181b. In contrast to older TargetScan predictions, the current parameters of TargetScan do not include any defined target binding site for the miR-204/211 cluster within the Meis2 3′UTR. Thus, current TargetScan predictions for Meis2 match the PicTar predictions regarding binding sites for miR-23a/b, miR-27a and the let-7/98 cluster, but also include miR-22, miR-30d and SC downregulated miR-30c (compare the Meis2 Web addresses in Materials and Methods and Figure 7B). Correspondingly, the potential competition of these other Meis2-specific miRNAs with miR-204-dependent regulation of MEIS2 expression, might account for the limited 15% downregulation of MEIS2 by pre-miR-204 in our experiment.
Including Meis2 and Runx2 TargetScan predicted a total of 322 mmu-miR-204-specific target mRNAs, suggesting that miR-204 regulation of other mRNAs might also be relevant for the observed SC attenuation (compare to the mmu-miR-204 Web address in Materials and Methods).
Bioinformatic Analysis of miR-184 and miR-204 Target Gene Ontology Database Reveals a Complex miRNA Network
To support the suggestion that miR-184 might regulate other SC-associated target mRNAs besides Bin3, the 18 TargetScan-predicted miR-184 targets were screened for a potential match with genes listed by the AmiGO gene ontology database (http://amigo.geneontology.org) under the GO terms GO:0001837:epithelial to mesenchymal transition and GO:0002088:lens development in cameratype eye. Besides Bin3, two other potential miR-184 target mRNAs, for example, eye development-associated factors Epb41l5 and Zic4, were identified by this method (60,61) (Figure 7A and Supplementary Figure 1A). Interestingly, transforming growth factor (TGF)-β-dependent upregulation of EPB41L5 participates in posttranscriptional regulation of cadherin and integrin important for EMT during mouse gastrulation. Thus, the role of EPB41L5 in SC-associated EMT in adult mice remains to be determined (62).
Similarly, the 322 TargetScan predicted miR-204 targets were screened for a potential match with genes listed by the AmiGO gene ontology database (http://amigo.geneontology.org) under the GO terms GO:0001837 and GO:0002088. Besides Meis2 and Runx2, nine potential SC-associated miR-204 target mRNAs were identified. For instance, the miR-204 targets Meis1, Wnt4, Foxc1, Hnrnpa2b1, Tfap2a, Mafg and Sox1 represent important factors in regulation of eye development (63, 64, 65, 66, 67, 68, 69) (Figure 7B and Supplementary Figure 1B). In addition, the two identified miR-204 targets Dlg1 and Tgfbr2 represent some potential candidates for regulation of SC-dependent EMT (70, 71, 72). For instance, upregulation of TGF-β2 receptors (TGFBR2) can be found in regenerating lens epithelial cells after UVB irradiation (72). In addition, DLG1 downregulation was associated with EMT in a mouse ocular tumor model (70).
Interestingly, when looking at expression of SC-associated miRNAs in correlation with binding to miR-204 target mRNAs, mRNAs of the major developmental factors, for example, Wnt4, Foxc1 and Hnrnpa2b, are solely regulated by miR-204/211 with two binding sites of miR-204/211 within the 3′UTR of Wnt4 (Supplementary Figure 1B). In contrast, 3′UTRs of Tfap2a and Sox11 seem to be coregulated by miR-204/211 and miR-92a. In addition, 3′UTRs of Meis2, Runx2, Tgfbr2, Dlg1, Meis1 and Sox11 share binding sites for the miR-23a/b cluster, suggesting a miR-204 counterregulatory function of miR-23a/b. Further common miRNA binding sites include the let-7/98 cluster in the 3′UTR of Bin3 and Meis2 and the miR-30c/d cluster in the 3′UTR of Epb41l5 and Zic4 as well as Meis2, Runx2 and Mafg. These findings further support existence of a competitive miRNA network for target mRNAs regulated by miR-184 and miR-204.
The authors declare that they have no competing interests as defined by Molecular Medicine, or other interests that might be perceived to influence the results and discussion reported in this paper.
This work was supported by National Institutes of Health grant EY16707 to PA Tsonis.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, and provide a link to the Creative Commons license. You do not have permission under this license to share adapted material derived from this article or parts of it.
The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
To view a copy of this license, visit (http://creativecommons.org/licenses/by-nc-nd/4.0/)